DE69706613T2 - Method of delivering uneven radiation - Google Patents

Description

The invention relates to a method for delivering a non-uniform distribution of radiation intensity from a radiation source to a field on an object.

Radiation emitting devices are well known and are used, for example, as a radiation therapy device for treating patients. A radiation therapy device usually comprises a gantry that can be pivoted about a horizontal axis of rotation during a therapeutic treatment. A linear accelerator is located in the gantry for generating a high-energy radiation beam for therapy. This high-energy radiation beam can be electron radiation or photon (X-ray) radiation. During a treatment, this radiation beam is directed at a zone of a patient that lies in the isocenter of the gantry rotation.

To control the radiation emitted toward an object, a beam shielding device such as a plate array and/or a collimator is usually provided in the trajectory of the radiation beam between the radiation source and the object. An example of a plate array is a set of four plates that can be used to define an opening for the radiation beam. The beam shielding device defines a field on the object to which a prescribed amount of radiation is to be delivered. A collimator is a beam shielding device that may include multiple blades (e.g., 50 blades). These blades are positioned to accurately direct the radiation beam toward the area to be treated with radiation. While these blades provide accurate alignment, they also allow for a small amount of unwanted radiation leakage. This leakage occurs between the blades.

The delivery of radiation by a radiation treatment device is prescribed and approved by an oncologist. However, the actual operation of the radiation equipment is usually performed by a therapist. When the therapist supervises the actual delivery of the radiation treatment as prescribed by the oncologist, the radiation emitting device is programmed to deliver that specific treatment. When programming the treatment, the therapist must consider the actual radiation output and adjust the dose delivery based on the plate array aperture to achieve the prescribed radiation treatment at the target depth in the subject. This adjustment can be made according to known calculations, but the therapist must perform it manually, which can lead to errors. In the context of radiation treatment, a miscalculation can result in either a dose that is too low and ineffective or too high and dangerous. A large error (e.g., a misplaced tens digit) can be fatal.

US 5 216 255 discloses a method for providing a radiation output to an array with a uniform intensity distribution by dividing the array into several sections.

It is an object of the invention to provide a method for efficiently and accurately delivering an irradiation treatment to a field having a non-uniform irradiation intensity distribution.

This object is achieved by a method according to claim 1.

Further developments of the invention are specified in the dependent claims.

Embodiments of the invention are described below with reference to the drawings, of which:

Figure 1 is a schematic representation of a radiation treatment apparatus incorporating a treatment console constructed in accordance with the present invention;

Fig. 2 is a block diagram illustrating portions of a radiation treatment apparatus and portions of a treatment unit in greater detail;

Fig. 3 is a three-dimensional illustration of a field to be treated with radiation;

Fig. 4 is another three-dimensional representation of a field to be treated with radiation;

Fig. 5 illustrates a process flow diagram for the present invention;

Fig. 6 illustrates a process flow diagram for a section optimization; and

Fig. 7A-C show matrices used for the section optimization process.

The invention is described below with primary reference to a system for delivering x-rays to a field on a patient and confining the field using at least one movable plate or jaw in the beam path from a radiation source. The invention can be used to control the delivery of any type of energy, for example electrons (instead of x-rays), to any type of object (not just a human patient), provided that the amount of energy delivered to the field can be sensed or estimated.

Fig. 1 shows a radiation treatment device 2 of a conventional design, in which plates 4, a control unit in a housing 9 and a treatment unit 100 are used. The radiation treatment device 2 has a frame 6 which can be pivoted about a horizontal axis of rotation 8 in the course of a therapeutic treatment. Plates 4 are attached to a projection (e.g. an accessory holder) of the frame 6. To generate the high-power radiation required for the therapy is needed, a linear accelerator is located in the gantry 6. The axis of the radiation beam emitted by the linear accelerator and the gantry 6 is designated 10. Electrons, photons, or any other detectable radiation may be used for the therapy. During treatment, the radiation beam is directed at a zone 12 on an object 13 (e.g., a patient to be treated and which lies in the isocenter of the gantry rotation). The axis of rotation 8 of the gantry 6, the axis of rotation 14 of the area on the object to be treated, and the beam axis 10 all preferably intersect at the isocenter.

The area of the patient that is irradiated is known as the field. The amount of radiation used to treat a field is specified in monitor units (mu). The plates 4 are substantially opaque to the emitted radiation. They are mounted between the radiation source and the patient to confine the radiation beam to approximately the shape of the field. Areas of the body (e.g. healthy tissue) are therefore exposed to as little radiation as possible, and preferably none at all. In the preferred embodiment of the invention, the gantry can be rotated to allow different beam angles and radiation distributions without moving the patient. This feature is necessary according to the invention. The invention can also be used with constant radiation delivery rates and with fixed angle beams (non-rotating gantry).

The radiation treatment device 2 also includes a central treatment processing or control unit 100, which is usually located remotely from the radiation treatment device 2. The radiation treatment device 2 is usually located in another room to protect the therapist from the radiation. The treatment unit 100 includes output devices such as at least one visual display unit or monitor 70, and an input device such as a keyboard 19. Data can also be stored by data carriers such as data storage devices or a verification and recording or automatic set-up system 102 described below. The treatment processing unit 100 is typically operated by the therapist who monitors the actual delivery of radiation treatment as prescribed by an oncologist. Using the keyboard 19 or other input device, the therapist enters into a control unit 76 of the treatment unit 100 the data defining the radiation to be delivered to the patient (e.g., according to the oncologist's prescription). The program may also be entered through another input device such as a data storage device, by data transmission, or using the automatic set-up system 102. Various data may be displayed on the screen of the monitor 70 before, during, and after treatment.

Fig. 2 shows portions of an irradiation treatment device 2 and portions of the treatment unit 100 in greater detail. An irradiation beam 1 is generated in an accelerator 20. The accelerator 20 comprises a gun 21, a waveguide 22 and an evacuated sheath or guide magnet 23. A trigger system 3 generates injector trigger signals and delivers them to an injector 5. Based on these injector trigger signals, the injector 5 generates injector pulses which are delivered to the gun 21 in the accelerator 20 for generating the X-ray beam 1. The electrons used to generate the X-ray beam 1 are accelerated and guided by the waveguide 22. For this purpose, a high frequency source is provided which supplies high frequency signals for generating an electromagnetic field which is supplied to the waveguide 22. The electrons injected by the injector 5 and emitted by the gun 21 are accelerated by this electromagnetic field in the waveguide 22 and exit at the end opposite the gun 21 as the electrons which will produce the X-ray beam 1. These electrons then enter a guide magnet 23 and are guided from there through a window 7 along the axis 10. After passing through a target 15, the electrons generate the X-ray beam 1 and this beam 1 passes through a passage 51 of a shielding block 50 and encounters a smoothing filter 17.

Next, it is sent through a measuring chamber 60 where the radiation dose is determined. Finally, an aperture plate assembly 4 is provided in the path of the radiation beam 1. The aperture plate assembly 4 includes a pair of jaws 41 and 42. As described above, this is merely one example of a beam shielding device that can be used in the invention. The invention will work with others as well, as long as there is an aperture plate assembly that defines the shape of an irradiated field.

The plate assembly 4 comprises a pair of aperture jaws 41 and 42 and an additional pair of aperture jaws (not shown) arranged perpendicular to the jaws 41 and 42. In order to match the size of the field to be irradiated, each of the aperture jaws can be moved with respect to the axis 10 by a drive unit 43 shown in Fig. 2. The drive unit 43 comprises an electric motor coupled to the jaws 41 and 42 and controlled by a motor controller 40. Position sensors 44 and 45 are also coupled to the jaws 41 and 42, respectively, for detecting their positions. This is merely an example of such a system. The invention will also work with other systems as long as there is a beam shielding arrangement defining the shape of an irradiated field and as long as sensors are provided to indicate the field size. For example, the plates can be replaced by a multi-leaf collimator containing many (e.g., 60) radiation blocking leaves. In the preferred embodiment, a collimator is used for the beam shielding arrangement.

The motor control 40 is connected to a dose control unit 61 which contains a dosimetry control and which is connected to a central processing unit 18 for providing set values for the radiation beam for achieving given isodose curves. The output of the radiation beam is measured by a measuring chamber 60. In response to the deviation between the set values and the actual values, the dose control unit 61 provides signals to the trigger system 3 which changes the pulse repetition frequency so that the deviation between the set values and the actual values of the radiation beam output is minimized.

In such a radiation treatment device, the goal of treating as much of the field as possible without irradiating healthy tissue is predominant. Fig. 3 is a three-dimensional illustration of a field to be treated with radiation. A three-dimensional intensity map 200 represents a possible field on a patient. The present invention divides the intensity map 200 into several intensity sections. Each of these sections is then individually treated with radiation. To effectively treat the intensity map 200, more, smaller sections are located near the contoured areas of the intensity map 200. For example, a large section (or rectangle in this case) 210 can be used as a base section to cover a large central section. The areas surrounding the large section 210 are then divided into smaller sections of variable sizes. A software program can be used to divide the field into a set of sections that can be effectively treated with the radiation. For example, the motion of the plates can be taken into account when dividing the intensity map 200 into sections. The sections can be more efficiently exposed to the radiation if plate rotation (or available plate motion) is taken into account. The three-dimensional aspects of the intensity map can also be taken into account. Thus, the software program optimizes the size and placement of each of the sections along with the order in which the sections are treated with radiation.

In one embodiment, the intensity map shown in Figure 3 can be divided into an equal number of monitor units by cutting the map. These static sections can then be reshaped and combined such that the sum of the monitor units does not change and efficient treatment results. In this situation, a matrix can be used to assist in dividing the field into sections. Instead of dividing and optimizing the simple intensity map in Fig. 3 can also be effectively treated as a complete region with a multi-leaf collimator. This subdivision and optimization allows the intensity map 200 to be treated with either (1) plates or (2) a combination of plates and a multi-leaf collimator.

Fig. 4 is another three-dimensional illustration of a field to be treated with radiation. The numbers in Table 1 are intensities at locations in space (within a field to be treated) identified in the x and y coordinate systems. Rectangular blocks 220-230 are used to represent the different intensities at these locations in space. By placing these separate rectangular sections 220-230 in a diagram, we obtain a columnar structure in Fig. 4. The height of each of the rectangular sections 220-230 is equal to the corresponding section intensity. Similarly, the positions of the rectangular sections 220-230 represent the corresponding location within the field.

Table 1 is a matrix identifying the sections in a field to be treated with radiation. All numbers in the tables below are monitoring units (mu). The four (4) in the upper left corner of the matrix (see Table 1) represents section 222 in Fig. 4. Similarly, the ten (10) in the lower right corner of the matrix represents section 230 in Fig. 4. Therefore, each number in the matrix of Table 1 represents one of sections 220-230. TABLE 1

Table 1 is used by the software program to prepare for treatment. In this example, the matrix in Table 1 is initially broken into two separate matrices. The first matrix is the section shown in Table 2. This is the largest possible section that can be obtained from the matrix in Table 1. TABLE 2

The second matrix is shown in Table 3. TABLE 3

When the matrices in Tables 2 and 3 are added together, the matrix in Table 1 results. The matrix in Table 3 is then broken down into the seven matrices shown in Tables 4 to 10. These matrices contain only 1s and 0s. TABLE 4 TABLE 5 TABLE 6 TABLE 7 TABLE 8 TABLE 9 TABLE 10

After the matrices have been broken down as shown above, the software program moves some of the 1's around so that some of the matrices can be combined. For example, if the 1 in the top row of the matrix in Table 6 is moved to another matrix, then the matrices in Tables 6 and 7 can be combined into the matrix shown in Table 11. TABLE 11

Therefore, the 1's can be moved around and the matrices combined to provide a more efficient treatment. In the preferred embodiment, the software program creates a set of static sections with a minimum number of matrices. The matrices are then organized into a specific order. This specific order takes into account the movement of the plates (or slats).

Figure 5 illustrates a process flow diagram for the present invention. At step 240, the software program begins. The parameters of the field to be irradiated are entered at step 242. Then the field is divided into intensity sections at step 243. As mentioned above, complex calculations are performed to efficiently divide the field into sections. Now the field is prepared for radiation treatment. The plates are positioned over the first section at step 244, and the first section is then treated with radiation at step 246. The software program then checks at step 248 whether all sections have been treated. If not all sections have been treated, the program returns to step 244. The plates are then positioned over the next section, and that section is treated with radiation at step 246. This is repeated until all sections have been irradiated. The software program ends at step 250.

Fig. 6 illustrates a process flow diagram for section optimization. Fig. 6 provides more details related to the optimization operation. In step 260, the software program determines the minimum radiation beam intensity, the shape of the field to be irradiated, and the leaf positions in the multi-leaf collimator. The minimum radiation beam intensity will define a section to be irradiated (e.g., see the matrix in Table 2). The leaf positions for the minimum intensity are needed to define the section shape. In step 262, the intensity map is treated as matrices. The uniformly filled minimum intensity matrix (e.g., see the matrix in Table 2) is subtracted from the initial intensity map. This provides a variation matrix (e.g., see the matrix in Table 3). In step 264, the variation matrix is divided horizontally into matrices with equal monitoring units. cut (e.g., see the matrices in Tables 4-10). The variation matrix does not have a uniform number of monitoring units throughout. This variation within the matrix is known as modulation. If an intensity map has a low modulation, the shape of the field changes little with increasing intensity. For example, (1) if there is a high modulation, one monitoring unit might be used, (2) if there is a moderate modulation, 3 monitoring units might be used, and (3) if there is a low modulation, 5 monitoring units might be used.

In step 266, each of the matrices resulting from the cutting is converted to a binary representation. This binary representation contains a 1 for the presence of radiation and a 0 for the absence of radiation. In step 268, the number of different plate or blade positions needed to handle the original binary matrices is calculated. For this calculation, multiple co-formed fields are treated as only one field. In step 270, the software program checks whether the theoretical minimum for the initial count of plate/blade positions has been determined. If not, the theoretical minimum is determined in step 272. If it has been determined, the software program moves to step 274. In step 274, a comparison is made between the number calculated in step 268 and the number determined in step 272. If the number of different plate/louver positions is greater than the theoretical minimum and the number of different plate/louver positions has decreased during the last 10 iterations, then the program moves to step 276. If the number of different plate/louver positions is equal to the theoretical minimum, then the program moves to step 278. Additionally, if the number of different plate/louver positions has not decreased during the last 10 iterations, then the program moves to step 278.

In step 276, the permutations of the 1s and 0s in the matrices are generated. These permutations are carried out according to a scheme chosen by the optimization routine of the software program. An example of an optimization routine is simulated annealing. Simulated annealing is a well-known optimization routine described in "Numerical Recipes in C" by Press, Teukolsky, Vetterling and Flannery, 1992, Cambridge University Press, pages 444-451. Simulated annealing may take a longer time (when compared to other optimization routines), but the results are the optimal configuration of the matrices.

In addition to simulated annealing, an approach that characterizes the intensity map slices in terms of plate/collimator rotations, "islands," same-shape slices, and/or intensity minima can be used. Plate/collimator rotations take into account the rotational positions of the stage. An "island" is an isolated shape in a binary matrix (i.e., a set of 1's without any 0's mixed in along a direction of leaf motion). For example, in Figure 7A, matrix 288 is broken into matrices 294 and 296 such that only one "island" is contained in each of matrices 294 and 296. Same-shape slices are created when 1's and 0's are redistributed such that the matrices become identical and can be combined (as shown in Table 11). Intensity minima are the smallest number of monitoring units in a variation matrix. For example, in Fig. 7A, the 1s in the upper left and lower right corners of matrix 284 are intensity minima. In this approach, after the intensity map slices (or the original matrices) are characterized, the intensity minima are moved around (e.g., to matrix 296 in Fig. 7A) until islands of common shape are found. Then, the next higher set of intensities is moved around, and the process repeats until the maximum number of islands of common shape is found. While this approach quickly produces a different configuration of matrices, these resulting matrices may not be the optimal configuration of matrices.

After step 276, the software program returns to step 268, which counts the number of plate positions required for the current configuration of the dies. The software program then goes to step 270, etc. In step 278, the plate/blade positions for each section and the sequence of the multi-section treatment. This determination is made to minimize plate/lamella movement.

Figures 7A-C show matrices used for the section optimization process. Turning first to Figure 7A, matrix 280 is a matrix representing the field to be treated with radiation. In this example of the section optimization process, matrix 280 is initially broken down into two separate matrices 282 and 284. Minimum intensity matrix 282 is the largest possible cube-like matrix. Variation matrix 284 is the second matrix containing the values left over after matrix 282 is removed from matrix 280. Matrix 284 is then broken down into three matrices 288, 290 and 292. These matrices 288, 290 and 292 contain only 1's and 0's. To begin the simplification process, matrix 288 is broken down into matrices 294 and 296.

In Fig. 7B, matrices 292 and 296 are combined to produce matrix 300, and matrix 288 is replaced by matrix 294. Matrix 284 now equals the sum of matrices 294, 290, and 300. In Fig. 7C, the upper left bit in matrix 294 (see Fig. 7B) is moved to the upper left position in matrix 300 (see Fig. 7B) to produce matrices 302 and 304. Now matrix 284 equals the sum of matrices 302, 290, and 304. Matrices 302 and 290 are then combined to produce matrix 306. Matrix 284 is now broken down into just two matrices 306 and 304. Overall, the original matrix 280 can be broken down into three matrices 282, 306 and 304. Each of these three matrices 282, 306 and 304 represents a section of the field. Therefore, in this example, the field is divided into three separate sections.

In the preferred embodiment, the plate position, irradiation dose rate and irradiation time are used to treat each of the sections with radiation. The plates are positioned over the respective section during treatment of the respective section. A multi-leaf collimator may also be used instead of or in addition to the plates. If a multi-leaf collimator is used, difficultly shaped matrices can be easily covered with the radiation beam. Ten 1 cm wide leaves could be used in the multi-leaf collimator. Unfortunately, multi-leaf collimators normally have some radiation leakage that occurs between the leaves. If the plates are used, this radiation leakage is avoided. Therefore, it is advantageous to use the plates without the leaves as much as possible. Certain section arrangements allow extensive use of the plates with minimal use of the collimator leaves. This occurs when most of the sections are rectangular in shape. In addition, radiation leakage is greater when the leaves move during treatment. In the present invention, the leaves do not move since they treat small static sections of the field. Complete coverage of the field can also occur without changing the rack position during treatment of a section. The stationary blades and gantry (1) eliminate the need for dynamic control electronics due to increased reliability of treatment delivery and (2) allow non-coplanar intensity modulated fields to be delivered to the field to be treated. These non-coplanar intensity modulated fields do not have to be in the same plane and they are characterized by a varying amount of monitoring units across the irradiation area. By allowing these fields to be non-coplanar and by allowing the intensities to vary across the field, healthy tissue will receive less radiation and sensitive organs can be missed.

The radiation delivered to an object can be analyzed into primary and scattered components. The primary radiation consists of the initial or original photons emitted by the radiation source, and the scattered radiation is the result of the photons being scattered by the plate array itself. The radiation output of the beam into free space increases due to the increased plate/collimator scattering added to the primary beam. In other words, a point in the field is exposed not only to the direct radiation (which is the primary component), but also to the radiation scattered by the plate array. Smaller fields have less radiation scattering. Therefore, when multiple sections are used to treat a field with radiation, scattering is greatly reduced.

Claims (14)

1. A method for providing a non-uniform distribution of radiation intensity from a radiation source to a field (200) on an object (13), the method comprising the steps of dividing the field (200) into a plurality of sections (222-230), each section having a desired total radiation intensity value, defining a plurality of dosage profiles, each profile assigning an associated radiation intensity value to a respective section such that the desired total radiation intensity value of each section is substantially equal to the sum of the radiation intensity values assigned thereto across all dosage profiles, and treating the object with radiation according to each dosage profile, wherein the treating step comprises the substeps of defining an opening between the radiation source and the object, the opening being capable of approximately confining a radiation beam for the dosage profile, and Generating a radiation beam from the radiation source to the object, the beam irradiating the object according to the dosage profile. 2. The method of claim 1, wherein each dosage profile defines a set of sections to be irradiated, all sections not in the set having negligible associated radiation intensity values associated with them. 3. A method according to claim 2, wherein each dosage profile assigns the same radiation intensity value to all portions of its defined set. 4. The method of any one of claims 1 to 3, further comprising the step of defining a radiation dose at the isocenter of each of the portions. 5. A method according to any one of claims 1 to 4, wherein the opening is defined by at least one plate, the plate being capable of blocking radiation from the radiation source. 6. The method of any one of claims 1 to 5, wherein the plurality of sections are static sections. 7. The method of any of claims 1 to 6, further comprising the step of using a first matrix (280) to represent the division of the field into multiple sections having separate desired total irradiance intensity values, and the step of using additional matrices (282-306) to represent the irradiance intensity values assigned by multiple dosage profiles. 8. The method of claim 7, further comprising the step of organizing the additional matrices (282, 304, 306) in a specific order. 9. A method according to any one of claims 1 to 8, wherein the opening is defined by a multi-leaf collimator, the multi-leaf collimator being capable of blocking radiation from the radiation center. 10. Method according to claims 1-9, in which an optimization routine is used to define a dosage profile. 11. Method according to one of claims 7-10, wherein an optimization routine is used to define a matrix order. 12. Method according to claim 10 or 11, in which an optimization criterion is minimizing the number of dosing profiles. 13. A method according to any one of claims 10 to 12, wherein an optimization criterion is finding sets of dosage profiles for which the sections with substantial radiation can be covered with a minimum number of rectangles that do not contain sections with negligible radiation. 14. A method according to any one of claims 10-13, wherein an optimization criterion is minimizing the adjustment changes required between successive dosage profiles.